The objective of this career development award is to provide Taylor S. Riall, M.D., Ph.D. formal postdoctoral training in healthcare policy, quality improvement, and leadership. This training will be coupled with mentored research on practice patterns, the utilization of surgical resection in patients with pancreatic cancer, and regionalization of care for this complex disease. This will enable Dr. Riall to make the full transition to independent investigator and leader in the field of health services research and health policy. A highly qualified team of mentors will supervise her training and ensure that she meets her goals. Dr. Riall will have resources available at the University of Texas Medical Branch at Galveston, the University of Texas M.D. Anderson Cancer Center, and the University of Texas School of Public Health. Pancreatic adenocarcinoma is the 4th leading cause of cancer deaths in both men and women in the United States. At the current time, surgical resection or removal of the tumor remains the only hope for long-term survival and cure for patients with this aggressive cancer. While clear evidence supports the use of surgical resection in patients with locoregional pancreatic cancer, recent studies by our group and others have determined that only 27-35% of patients with potentially resectable locoregional disease undergo surgical resection. In addition, only 75% of patients with locoregional disease are evaluated by a surgeon. Dr. Riall will use the SEER-Medicare linked data and data from the American Medical Association Physician Masterfile to achieve the following specific aims: 1) To describe the evaluation process of patients with locoregional pancreatic cancer including the percentage of patients who see a specialist and the percentage of patients who receive key diagnostic and staging tests, 2) To describe the patient, tumor, surgeon, and medical system characteristics which predict the receipt of surgery in patients with locoregional pancreatic cancer for all patients evaluated by a surgeon, 3) To evaluate the extent of and trends in regionalization of care to high-volume centers (>10 pancreatic resections per year) among Medicare patients with locoregional and metastatic pancreatic cancer, and 4) To assess variability in care among Medicare providers performing pancreatic resection to determine if volume alone is an appropriate criteria for pancreatic surgery referral. This unique, systematic, population-based analysis of the variation in use of surgical resection for pancreatic cancer at the levels of the patient, tumor, provider, and medical system will enable us to determine the reasons for underutilization of surgical resection. These data can be used as a basis for developing clinical practice guidelines and health policy for patients with pancreatic cancer, thereby reducing variations in practice patterns and optimizing outcomes for patients with this aggressive disease.